As an alternative to sintering, the melt infiltration method can be used to create a dense component from a porous molding body. The prerequisite is that the base material consists of a porous body having a higher melting point than the infiltrating material. Furthermore, the melt must wet the base material. Then the porous body and the infiltrating material can be heated up, until the melting point of the infiltrating material is exceeded. The melt is drawn through the pores of the body by capillary forces and completely fills the pore volume. After cooling down, you will achieve a dense component, with – if correctly processed – the exact dimensions of the molding body.
By melt infiltration, ceramics, metals and metal-ceramic composites can be produced. The best-known composite is SiSiC ceramics, which is produced by infiltration of silicon melt in porous bodies made of silicon carbide. The so-called LSI process (Liquid Silicon Infiltration) is also used to infiltrate porous bodies made of carbon or SiC fibres, whereby CMC components (Ceramic Matrix Composites) are produced. Other hard substances can also be compressed by melt infiltration. Thus metallic binders such as cobalt or nickel can be infiltrated into porous bodies made of tungsten-carbide to manufacture hard (carbide) metals. One area in the field of oxide ceramics is e.g. the infiltration of glass melt in a porous body made of aluminum oxide. The glass melt crystallizes as it cools and you obtain a largely crystalline material. One area in the field of metals is e.g. the infiltration of bronze in porous bodies made of steel.
The melt infiltration process facilitates the design of microstructures that are often impossible to achieve by sintering. Reactive components, for example, can be introduced into the porous bodies before melt infiltration. During infiltration, new phases are formed due to the reactions with the melt. For example, during the LSI process additional carbon can be introduced into the porous bodies. During the infiltration this carbon reacts with the silicon melt forming additional silicon carbide, significantly increasing the hardness and stiffness of the SiSiC material. The thermal expansion coefficient of the involved phases can be adapted to one another so that stresses are reduced. Even gradients in the material composition and the material properties can be achieved via the reactive melt infiltration process. The methods are particularly suitable for the microstructure property design that has been developed at the HTL center.
Porous bodies for melt infiltration can be manufactured with typical powder metallurgical shaping processes like dry pressing, injection molding or extrusion. Green bodies must be debinded before melt infiltration. As the porous bodies can already be produced in the final shape, forming processes that facilitate the production of more complex shapes are interesting. This includes 3D printing, in particular. Just as with sintering, the quality of the green and debinded body directly influences the quality of the infiltrated components. The HTL has specific measurement methods available for the evaluation of the green body quality.
The parameters for the melt infiltration process must be carefully optimized in order to finally achieve a product of high quality. This is done at the HTL by a combination of in situ measurement and finite element (FE) simulations. The kinetics of the melt infiltration are measured using a special ThermoOptical measurement system (TOM) in the relevant furnace atmosphere. The increase in weight of the porous body is monitored during infiltration. The temperature changes that occur in particular during reactive melt infiltration processes are measured simultaneously. The wetting properties of the melt with regard to the base material can also be evaluated with help of the TOM system. The heat transfer characteristics of the porous body are measured as a function of temperature before and after infiltration using Laser-Flash technology. All data is compiled in an FE model especially developed for the melt infiltration process. Using the FE model, the melt infiltration process can be simulated and optimized. In particular, the heat management is essential for success. The thermal effects that can occur not only during heat up, but also during cool down can deteriorate the quality of the infiltrated components, e.g. thermal stresses. The complexity increases significantly with the size of the components. Using the FE model developed at the HTL, the properties determined from small samples can be transferred to large components.